Comets falling into their star? Spectrometric observations have shown that comets constantly fall into their central stars. Here is p Pictoris, in the southern sky, as seen by the 3.6-metre telescope of the ESO. The star is surrounded by a disk of gas and dust. This disk is certainly evidence of the intense activity associated with young planetary systems. The image of the central star has been masked using the coronagraphic technique, allowing us to see the disk at less than 25 AU from the star.
others. The abundance in stars of chemical elements heavier than helium (which astronomers collectively and idiosyncratically call 'metals') seems to be a key parameter. The detection rate climbs by a factor of 7 in studies of stars with twice as many heavy elements as there are in the Sun. This is not a result of some observational bias, as stars of metallicity equivalent to the Sun's are much more common than those with more metals.
Only a handful of stars with lower metallicity than the Sun's are known to have planets. What can this result mean? We know that terrestrial planets are made of heavy elements: silicon, oxygen, nickel and iron. The giant planets which we observe orbiting other stars are undoubtedly gaseous, but their hydrogen-helium atmospheres probably conceal cores rich in heavy elements. These are therefore an essential ingredient in the formation of planets as we know them. It seems likely, then, that planets form more readily in environments rich in heavy elements - elements also found in the atmospheres of their stars, which were born of the same material.
However, we also know that planets can (and, indeed, must) migrate towards the interior of planetary systems. The more readily the process of migration begins - perhaps by interaction between planets and the protoplanetary disk -the less probable it is that the migration can be halted. Could it be that some of those planets situated at just hundredths of an AU from their stars end up by plunging into them? If this is so, then the rocky material of those planets will be vaporised in the star's atmosphere. But this cannibalism of planets does not seem to be the dominant process. On average, there is 1.8 times as much iron in a star accompanied by planets than in a star with none. If we calculate how much rocky material is required to explain this difference, we realise that stars would have had to absorb tens of planets. Now, observations confirm that protoplanetary disks have a lifetime of less than ten million years. The swallowing up of just a few planets by a star might occur in such a short period, but it is much less likely that tens of planets could suffer the same fate during that time. We have learned an important fact: that planets form more readily in an environment rich in metals. Models simulating the formation of planetary systems will have to take this into account. It will also be necessary to examine the possibility of the formation of planets in environments of medium metallicity - the best example being our solar system, as the Sun resides in the range of metallicity where the probability of finding planets is quite low!
Was this article helpful?